Banks have big balance sheets, and they need to do something with this capital. The role of the proprietary trader is to make money for the bank ‘ but often times, he’s out the door as soon as he has a bad year. Julie Ros asks whether banks are applying the right performance measures to these dealers, or do they have to live up to unrealistic expectations? <?xml:namespace prefix = o ns = “urn:schemas-microsoft-com:office:office” />
Like all areas in the dealing room, proprietary trading has undergone the usual up and down cycles in terms of its popularity among banks. With the introduction of the euro, the Internet and mergers among banks and even some customers, it appears that banks are once again turning to proprietary trading as a means of replacing lost revenues. But are banks setting themselves up to repeat past mistakes, or is there a way that proprietary trading could represent a steady revenue stream?
The planned introduction of FXall ‘ the online bank to client Internet portal set up by seven leading international banks and investment banks ‘ stands to change the way banks generate revenues and decipher flow information. ‘FXall is a major deflationary force. With machines, you can’t decipher the information anymore. The largest customer banks still can, but the situation isn’t likely to improve. The only way to circumvent this inflationary spiral is to maintain customers,’ says one proprietary trader. ‘If you allocate to a CTA, you have a captive customer, but you also need in-house prop traders to manage risk.’
’As the FX business consolidates ‘ be it into a single currency or onto Web-based dealing platforms ‘ banks must look to additional sources of revenue,’ adds Richard Breslow, head of global FX trading for Bank of America. ‘Technology enables us to keep in-house, a lot of the information about cross border and directional flows, which means we can capitalise on these directional moves rather than rely solely on the bid/offer spread. Banks need to make the most of their resources and be smarter about how they use this information.’
BofA is one of several banks that have lately been focusing on proprietary trading. The bank recently set up a global proprietary unit under Mark Brown, who joined the group at the end of March from HSBC. ‘In the past, we have had individual traders operating across the three centres, but this new group will take a more co-ordinated approach to their portfolio. We’re looking for people with complementary skills to create a real portfolio of risk,’ Breslow says.
Many banks had hugely successful proprietary trading operations in 1993, but after the bond market collapse in 1994, they became increasingly conscious of risk adjusted returns. Proprietary trading came back into fashion a couple of years later, only to be scaled back again following the 1998 Russian default.
Presently, proprietary trading appears to be enjoying a bit of a resurgence. JP Morgan just recently reported success from its proprietary trading activities, which soared during the second quarter of this year. The bank earned $283 million from proprietary trading, up from just $23 million for the same period last year.
’Traditionally, proprietary traders have made their money from trending markets, yet for the last 18 months, we haven’t seen the big percentage moves. Even currency traders at hedge funds ‘ reputedly a sector responsible for up to 60% of global FX turnover prior to the emerging market crises of 1997/98 ‘ are effectively sitting on their hands,’ says Mike Goggin, director at executive search firm Brookleigh Services.
’Today there are banks adding to their proprietary trading capabilities, there are banks de-emphasising this function, and there are banks considering entry,’ he adds. ‘In almost all cases the rules have been re-written. Presently, comfort to employers comes from traders that can make $5 million per year in current markets. The question is whether currency traders who have made $20 million or more in years gone by have been able to scale back. Can they now be turned on by 70 pips as opposed to 7 per cent?’
’One thing is for sure, it doesn’t matter who you are or how big you are, making money trading FX is more difficult than it has ever been,’ says Goggin.
The issue of pay among proprietary traders can be a contentious issue. Many prop traders argue that they should be compared to their ‘peers’, whom they consider to be the fund managers, not the market makers on the spot desk.
’It is very important to make a distinction between flow trading and directional trading ‘ because some banks blur this distinction,’ says another prop trader. ‘These banks sit their prop traders with the spot desk, which means these are actually flow traders whose returns should be compared to the market makers. True prop traders should sit separately, and therefore be compared to the performance of hedge funds, futures funds or fund of funds. Unless you do this, you’re comparing apples to oranges. When banks do this, they often reach the conclusion that they do not want prop trading within the group, saying that it’s too volatile compared to market making. This is why setting up a separate entity makes much more sense.’
Many prop traders believe risk adjusted returns are the only way to measure their performance. Some even believe banks should sign contracts with the prop desk, setting out risk adjusted performance measures in black and white.
’It’s liquidity management at its best, because prop trading is based on risk adjusted earnings, so the shareholder can’t lose,’ says another trader.
The issue of benchmarking is critical, say these traders, because when you trade as a hedge fund manager, the return to risk ratio is much lower than it is in market making.
In a working paper issued last year by the Federal Reserve Bank of Chicago, Product Mix & Earnings Volatility at Commercial Banks: Evidence from a Degree of Leverage Model, by Robert DeYoung and Karin Roland, the researchers explain why many banks feel uncomfortable with certain trading activities and therefore feel they must reduce Value at Risk (VAR).
The authors tested the degree to which revenue volatility is associated with the product mix of five revenue shares – deposit activities, lending activities, investment activities, fee-based activities, and trading activities – and control variables for banks’ size, merger activity and a constant term.
In the table, the authors test the degree to which the cross-sectional variation in revenue volatility is associated with product mix. Trading activities have an unambiguous positive effect on revenue volatility, says the report. Using the point estimates from column , moving one percentage point of revenue out of loans and into trading activities would increase revenue volatility at the average bank by 0.0124, or by about 11% (0.0124/0.1075). Bank revenues were much less sensitive to changes in the other four activities.
“The reason you see more volatility with pure proprietary trading in this example is because it does not include the smoothing effect of bid/ask spreads charged to clients,” notes the first trader.
Following the Russian debacle in 1998, Barclays reduced its proprietary trading activities in 1999 to lower VAR within the bank. The bank issued a statement that said, “Overall dealing losses within non-customer related proprietary trading businesses (including Russia) for the full year totalled £205 million, the bulk of which was incurred in August 1998. In addition, the provisions for bad and doubtful debts by Russian counterparties in currency forward contracts and repurchase agreements was £130 million out of a total charge of £159 million for the year. The country transfer risk provision charge was £10 million. In response to the deterioration in market conditions experienced during late summer and early autumn, management took a number of steps to improve business performance. The non-customer related proprietary trading businesses were closed in October and secondary market corporate bond inventory was substantially reduced during the last three months of the year. This overall reduction in risk appetite has resulted in a fall in weighted risk assets and total assets in the second half of the year of 18% to £31 billion and 23% to £117 billion, respectively.”
Some proprietary traders believe that this view supports the need for benchmarks that pre-agree a suggested Sharpe ratio of at least 1.0 over a three-year period. “It’s not surprising that banks feel proprietary trading is more risky than spot trading when it is compared with market makers rather than other investment classes such as the S&P or Nasdaq,” says the second trader.
John W. Henry & Company describes the Sharpe ratio as a measure of risk adjusted returns that is calculated using a programme’s excess returns (its return in excess of the return generated by risk-free assets such as Treasury bills) divided by the programme’s standard deviation. The standard deviation measures the variability of a probability distribution and is widely used as a measure of risk.
“Funds generally have between a 1-1.5 return to risk ratio. Proprietary trading levels exhibit similar returns to risk. So unless banks are prepared to agree a return to risk ratio of 1.0 compared to funds, they should not invest in prop trading,” says another prop dealer. “In the banking environment, benchmarking is crucial because you must compare like with like.
Understanding the stop-loss is key. The stop-loss is determined by volatility and the size of the position, and should be adjusted over the life of a trade.
Many treasurers only give stop-losses to prop traders, adds another dealer. “Stop-losses within the trading desk in dollar terms can be very confusing. There is a tremendous lack of transparency in this area. Bank managers need to understand the right ratio, target (budget) to stop losses – which again comes back to the issue of benchmarking,” says the dealer.
“A passive investor may achieve an annualised return of, let’s say, 40% by investing in the Nasdaq, but to realise such a result, he will have to hold the stocks, experience periods of exceptional volatility and potentially suffer large drawdowns,” the dealer adds. “Inserting a yearly stop-loss of 10% will significantly lower the expected return. This is due to the significant probability of being stopped (which is actually calculated when pricing knock-out options).”
“Senior management should be aware of this when setting targets,” the dealer continues. “Realistic budgets divided by stop-loss need to be pre-agreed and discussed at the earliest possible stage. Otherwise, the temptation to “trade the traders” as soon as he loses or makes money may soon appear. Enforcing a clearly defined stop-loss could therefore constitute a determinant step in the success or failure of a proprietary trading desk. If badly set-up, this will result in the rapid closure or downsize of proprietary trading activities. If rightly fixed, this will reduce risk and could actually improve the return to risk trade-off. Once more, a return to risk ratio may be more relevant to accurately value traders.”
The argument that prop traders should be measured the same way as hedge fund managers can be applied to bonus structures as well. If a prop desk has four traders and at the end of the year, two have performed well but two have lost money, leaving the desk flat, how do you compensate the two traders that made money?
“You don’t have that problem with hedge funds, because you just charge a 20-25% performance fee,” says the dealer. “If bonuses for prop traders are to reflect those of our peers in the hedge fund industry, bonuses should be contract related for each individual trader.”
“Most banks want the prop desk to make lots of money, but as soon it loses money, the game changes. Spot trading on the other hand, fulfils a franchise role – if they make money, the corporate client does as well. The costs of running spot and prop desks are the same, but the costs of having spot traders are justified by the corporate business,” the trader adds.
Increasingly, banks have begun outsourcing their risk taking activities to outside fund managers and CTAs (commodity trading advisors). When banks outsource to fund managers, not only do they obtain a captive customer base, but they are also privy to their positions – which is valuable in terms of flow information. But there is another aspect to outsourcing – that of compensation. Instead of paying the high base salaries plus bonuses to prop traders, funds are paid only on performance.
“There is a cultural hurdle that banks need to make in terms of outsourcing the proprietary business to a fund manager,” notes Alan Eisner, co-managing director at Millennium-T. “It makes sense to outsource some proprietary risk to organisations with the breadth and depth to generate P&L that is compatible with the organisation’s risk outlook. And, if banks outsource to a CTA, they gain a captive customer, because it’s often a condition that the CTA can only trade with the allocating bank.”
“In 1995-96, a lot of banks went too far in reducing risk, and went into a risk reduction mode after the 1994 bond market debacle, ” says Michael Huttman, managing director at the recently merged fund manager, Millennium-T. “We satisfy banks in terms of outsourced proprietary capital. Not only can we make them good returns, but we will take mandates on an either an unfunded or VAR basis, so banks don’t face capital requirements. Banks can also use us as a benchmark for judging how their own prop desks are performing.”
A few banks have actually taken the idea of outsourcing a step further, setting up their own internal hedge funds, which source capital externally.
UBS has historically had very active currency funds, dating back to the days of Swiss Bank Corp and O’Connor & Associates. On 1 June of this year, UBS Asset Management launched a new unit called UBS O’Connor, which comprises part of the former proprietary equity trading group, as well as the Fund of Funds and Currency Funds businesses of UBS Warburg. Chicago-based O’Connor focuses on alternative investment strategies designed to provide attractive risk-adjusted returns that have low correlation with traditional investments.
Meanwhile, Allied Irish Bank took the unusual step seven years ago, setting up a hedge fund/CTA, Allied Irish Capital Management Ltd (AICM). AICM currently has in excess of $1 billion under management, and is overseen by founding member Gerry Grimes.
Following its merger with NatWest Financial Markets, Royal Bank of Scotland Financial Markets is also taking steps in this direction, setting up an external hedge fund under Howard Kurz, which will cover multiple asset classes.
When banks do decide to launch their own hedge funds, there are obviously certain considerations to be made. “Banks can be very helpful in terms of control, compliance and management, but in terms of the day to day operations of the company, I don’t necessarily believe that a bank is a natural habitat for a hedge fund/CTA operation,” says AICM’s Grimes. “In order for the business to operate successfully within that environment, it has to be removed from the mainstream. That means being given a lot of autonomy and being allowed to structure itself as closely with its peers as possible.”
“Part of our success is due to the fact that the bank has developed a high degree of trust, and has allowed us to develop the business without much interference,” he adds. “We operate as a business within the banking group, which manages money for outside investors, much like most other areas in the trading room.”
Hub of InformationSome proprietary traders believe that prop trading could actually form the hub from which the entire dealing room operates. “With the popularity of the Internet, clients increasingly want more and more of a bank’s proprietary information available for free – from research and economic reports, to pricing and back office services. Banks are overloading the boat with costs, so how do you fund this?” says one senior prop trader.
Proprietary trading should be at the centre of a whole hub of information, says the trader. “It’s really a combination between trading and sales, because if the bank is taking a position, it’s a lot more convincing to make suggestions to your customers if the firm has put its own capital behind the strategy,” the trader says.
“The proprietary desk can generate ideas that can be exploited by the rest of the organisation – in terms of both sales and structured products,” adds another trader. “The prop desk can launch new financial products and generate meetings with clients – both of which make a suggested trade much more credible if the bank is taking the positions itself.”
If set up correctly, these traders believe the prop desk can provide a value-added service for clients. “If it can establish a track record, you improve your credibility with clients and so can market this to them,” says the trader.
Finding The Right Balance
Others believe that it is a combination of both in-house prop traders, outside fund managers and internal bank hedge funds that will form the best means of running proprietary trading businesses. But this, they say, should be run as a single unit.
“There are many different models, but at the end of the day, it’s the people that will determine whether it works or not,” says Grimes. “There is a bit of innocence out there about making the transition from proprietary trading to managing peoples’ money. There is a misconception that it’s easy to translate a good trader into someone good at managing investors’ money. It is a different set of psychological pressures that is involved in managing money for clients versus managing a bank’s capital. I think that’s why the model hasn’t really worked to date.”
“Last year marked the bottom of a cycle for prop traders,” adds another prop trader. “But I don’t know if banks will be smart enough to catch on and build long-term, successful businesses because banks would have to become ‘honest’ about the way they pay traders. If they don’t pay these traders like a hedge fund, they will continue losing their best traders to funds.”